Hello,
No space left on device, mysql error 28 (if mysql backup is enabled) or others interrupt backup system.
Last try:
“Servers to save” “Only selected”
“Features to backup”
“Only those selected below”
“Check all”
Unchecked both “Server’s home directory and web pages ( Include logs directory)”
Unchecked Contents of server’s MySQL databases
Also unchecked “Include sub-servers of those selected?”
“One file per server (old format)”
‘Copying Apache log files …’ Why is this if I asked not to include logs dir? (It creates a 2Gb file in tmp dir)
Restoring what I could backed up I got
"Extracting backup archive file …
… extraction of /…/domain.com.tar.gz failed!
/bin/tar: domain.com_web_alog: Wrote only 4096 of 10240 bytes
/bin/tar: Skipping to next header
/bin/tar: Archive contains obsolescent base-64 headers
/bin/tar: Error exit delayed from previous errors
… failed! See the progress output above for the reason why."
Many inconsistences.
Backup WITHOUT mysql data
Pre-Restore Report
(Server does not exist, will be re-created)
Virtual server password, description and other details
Webalizer configuration and schedule
Mail/FTP users and mail aliases
Logrotate configuration for log file
Apache webserver configuration
Restoring… a clash was detected : A MySQL database named *** already exists
… failed! See the progress output above for the reason why.
It seems to be a bug indeed that it copies the Apache logs even though you turned that feature off. I just tried that and can confirm it, both with “old” and “new” format. You might want to file that as a bug report so the developers see it.
Aside from that… 2 GB for Apache logs? Don’t you use log file rotation? The backup function seems to copy only the latest logfile, not archived ones.
The rest of your issues might well be caused by that no space left on device you’re seeing there. Is your HDD full?
About “no space left on device” I had this issue last server move about 18 months ago.
So I opted to use rsync for dirs. There is and there was enough HDD room, 150Gb+
Yes, there is log rotation. gz files are around 180MB.
Thanks.
Looking inside virtualmin logs I found a domain that hasn’t logrotation.
I don’t know why, it is an old one.
How can I add logrotation for access and errors? I thought there was an option in Edit Server.
Thank you.
If you look in /etc/logrotate.conf, do you see a mention of that particular account or domain?
You may also want to run the logrotation manually from the command line, so that you can see if there’s any errors. You can do that by running this as root:
Thanks, Eric. Logrotation is set. I thougt it wasn’t bacause /var/log/virtualmin doesn’t contain *n.gz rotated files. It seems they were not moved. Probably my fault
Locutus: Uncompressed weekly access log file is 2500Mb (180Mb, compressed)
You might - aside from the aforementioned bug that should sure be fixed - consider rotating more often then, if disk space is such an issue that Vmin cannot create a file in /tmp large enough to hold your 2.5 GB log.
Disk space is not a problem. Anyway I had another 8Gb weekly log file so I turned it to daily.
That domain has several Gbs in /home dir so I decided to use rsync to move data to new server, same as log file.
This is what I never understood. A year and a half ago I had the same problem 200Gb+ free space and 15Gb VS size and couldn’t perform a complete backup because of this.
